The Burhān by Mohammed Hijab

In this chapter, I will assess the responses of some Muslim and Christian theologians to Ibn Sīnā’s arguments. Among the Muslim theologians, I will refer to the writings of al Ghazālī, Ibn Rushd, and Ibn Taymiyyah, all of whom had intriguing responses to Ibn Sīnā. Many of these scholarly responses to Ibn Sīnā’s arguments can be accessed in English, with the exception of Ibn Taymiyyah. The latter’s Kalām-based arguments for God’s existence are widely untranslated or disregarded within academia. His contribution is particularly valuable, as he deals not only with the theoretical elements of the argument, but also their pastoral and apologetic applications. Moreover, although the Muʿtazilites engaged actively with the theological area of proving God’s existence, it appears that the major argument they employed was dalīl al-aʿrāḍ wa ḥudūth al-ajsām, which translates to ‘the proof from accidents and commencement of bodies’.29 This specific argument goes beyond the remit of this book, and requires a separate treatment altogether. Finally, I will then assess some of the uses of Ibn Sīnā’s Burhān in the Christian tradition, with particular reference to Thomas Aquinas and John Duns Scotus. The most important points of extrapolation from all of these thinkers are reformulations or criticisms of the argument. These can be considered as a means to strengthen the argument, refine it, or even add to it. Despite the objections that many of these thinkers raise with certain formulations of Ibn Sīnā’s argument, practically all of them converge on his conclusion of the reality of a ‘necessary existence’, which explains all other forms of existence. Finally, this section argues that the conclusion of a ‘necessary existence’ is so pervasive within interreligious intellectual circles that it may very well be the most agreed upon belief between all monotheists in the history of theological philosophy.

Ibn Sīnā’s Burhān argument goes as follows. There is existence. Existence is of three types: possible (contingent), necessary, and impossible. Impossible existences (like squared circles) cannot exist. There cannot be only contingent existences in existence, as they would require something else in order to cause them into existence. There cannot be a finite or infinite series of contingent existences, as such a series would be composed of many differentiated and dependent members. The differentiation and dependent aspects found in the different members of the series indicate that the finite or infinite series itself cannot be necessary. Instead, it must be contingent. Thus, only a necessary existence can cause or ultimately explain why any contingent existences happen to occur

Al-Ghazālī was perhaps the most famous critic of Ibn Sīnā in the Muslim world, with his publication of Tahāfut al-Falāsifah (The Incoherence of the Philosophers), where he directly attacked Ibn Sīnā’s beliefs. Al-Ghazālī’s criticism was so scathing that it led him to excommunicate Ibn Sīnā for four different reasons outlined at the end of his book, the most relevant of which surrounds Ibn Sīnā’s views on the eternity of the universe. In what will follow I will outline the critical differences found between the approaches of Ibn Sīnā and al-Ghazālī in demonstrating God’s existence. Particular focus will be given to the subjects of infinite regress and eternality, as well as arguments against emanationism and Godly will. Though al-Ghazālī does not directly challenge any of the premises of Ibn Sīnā’s overall argument mentioned above, nevertheless his criticisms of Ibn Sīnā’s general approach are just as important. After all, they give the theistic apologist more argumentative options than would be afforded by just relying on Ibn Sīnā’s articulations of the Burhān alone. As it relates to the infinite regress, al-Ghazālī makes the argument that ‘anything susceptible to greater or lesser is finite’. This is concurrent with the proofs already employed by the likes of John of Philoponus.30 Ibn Sīnā was an eternalist (believing the universe is pre-eternal), as well as an emanationist (believing that the universe emanates from God necessarily). Expressed in another manner, Ibn Sīnā believed that God’s bringing of the universe into existence is analogous to the Sun emitting light. In this simile, the light represents the universe and the Sun symbolises God.31Al-Ghazālī, however, considers eternalism to be inconsistent with an omnipotent God, as it renders him impotent to prevent things from coming into existence, or from not coming into existence in the first place.32 Indeed, al-Ghazālī (alongside other Ashʿarites who preceded him) made arguments from ‘particularisation’ which aim to establish the divine will through cosmological proofs. Within this context, they give examples like those of rotating planets in space that had the potential of rotating the other way.33

Exploring the Ashʿarite argument from particularisation is important, as it is almost totally absent in discourses in the philosophy of religion (as an argument for the will of God), despite the high-level scope the argument holds. Unfortunately, it could be said that the Ashʿarites were not well-suited to deal with possible determinist objections (to be dealt with in the chapter on objections). This is because many of its key thinkers, including al Ghazālī, were occasionalist in inclination, thereby anticipating David Hume by arguing that the connection between cause and effect is not necessary. Oftentimes, they use the example of fire that seemingly causes combustion to cotton, while in reality arguing that causation is from God directly.34 Occasionalism, of course, weakens the first premise of al-Ghazālī’s own argument, which states that: ‘Everything that begins to exist has a cause, the universe began to exist, therefore the universe has a cause’.35 This is because the premise ‘everything that begins to exist has a cause’ cannot be established inductively if al-Ghazālī actually argues – in other works – that causes are directly from God. If these causes are actually what are being referred to, then al-Ghazālī may be accused of begging the question. In refutation of this viewpoint, the Ḥanbalite scholar Ibn Taymiyyah argues that secondary causation does not endanger the independence of God, namely by 31 Leaman, O. (2000). A brief introduction to Islamic philosophy. Polity Press, p. 41.

32 Proofs for eternity, creation and the existence of God in medieval Islamic and Jewish philosophy, p. 4; Griffel, F. (2009). Al-Ghazali’s philosophical theology. Oxford University Press, p.185.

establishing the difference between ‘direct causation and indirect causation’.36 Ibn Taymiyyah reasons that things in the world are contingent in and of themselves and only necessary because of their connection to the necessary existence.37 This is to say that things in the universe are contingent in abstraction, and determined in their connection with the necessary existence. Although New Atheists (and other non theists) have not made this formal objection to the contingency argument in the literature, it can be anticipated that the determinists among them would take such a course of action. A key example in this regard may be Sam Harris, who makes the case for determinism in his book Free Will. On this line of reasoning the status of an object as ‘contingent’ will be put into question on account of it being determined by antecedent causes, where it could not be in any other way. By differentiating things that could not be any other way in abstraction (in and of themselves) from those things that could not be any other way because of another thing (say an uninterrupted causal chain or an uncaused determiner), one can circle this objection.

Another subject of enquiry is al-Ghazālī’s suspicion of Ibn Sīnā’s modal categories as being relevant to the real world. Regarding this, Griffel states that ‘Al-Ghazali questions the assumption of an ontological coherence between this world and our knowledge of it. Certain predications – which, for Avicenna, apply to things in the real world – apply, for Al-Ghazali, only to human judgements’.38 This ‘suspicion’ is concurrent with the Kantian objection to the ontological argument. It cannot be reasoned, however, that al-Ghazālī did not believe in possibility and necessity as existing in the ‘real world’. For he and the Ashʿarites that preceded him made an argument from particularisation, as explained by Griffel: The idea of particularisation (takhsis) implicitly includes an understanding of possible worlds that are different from ours. The process of particularisation actualises a given one of many alternatives. Yet the alternatives to this world – Which would be: “X comes into existence at a time different from when X comes into existence” are not explicitly expressed or imagined. The Kalam concept of preponderance (tarjih), however, explicitly discusses the assumption of a possible worlds….39

Griffel goes on to postulate that al-Ghazālī’s version of this particularisation has ‘strong overtones of Avicenna’s ontology: because everything in the world can be perceived as nonexisting, its nonexistence is itself equally possible as its existence. Existent things necessarily need something that “tip the scales” its existence…God is this preponderator (murajjiḥ) who in this sense determines the existence of anything in the world’.40 The argument of particularisation is of great significance, as it allows theists to make an argument for a volitional God. Such a feature is not available through a Burhān argument alone and is therefore invaluable to any theistic apologist.

Ibn Rushd’s critique of Ibn Sīnā’s argument presented above does not relate much to the flow of the argument. Instead, it largely lies in the way Ibn Sīnā defines the modal categories, especially the category of mumkin al-wujūd (possible or contingent existences). In his famous invocation of Aristotle, Ibn Rushd mentions that ‘it is impossible for any science to demonstrate the existence of its own subject matter’.41 In his Pointers, Ibn Sīnā defines possible (contingent) existence as something which is ‘not impossible and not necessary’.42 In addition, he indicates that if such a contingent existence is not in existence no logical absurdities would occur. Ibn Rushd claims that in order for Ibn Sīnā’s Burhān argument to be properly made, an appropriate definition of possible existence or mumkin al-wujūd must be adopted, namely ‘what is generated or destroyed’. This must be the analytical starting point, for it is empirically attested.43 If one does not start with this kind of definition, one will be at risk of equivocating between that which is contingent and that which is caused. In fact, Ibn Rushd accuses Ibn Sīnā of equivocating between causes and contingencies; within this critical discussion he states that contingent things are broader than causes.44 Importantly, Ibn Rushd accuses Ibn Sīnā of claiming that contingent things are caused and that in his view this distinction ‘is not a division that considers the qua existent’.45

Bearing Ibn Rushd’s criticism in mind, in my own presentation of arguments I will make sure that I differentiate between causes, dependent things, and contingent things. Thomas Aquinas and the Third Way Ibn Sīnā’s Burhān was influential to the degree that it appears in the works of Thomas Aquinas.46 In addition to Ibn Rushd, the approach of Aquinas is commensurate with that of his Ḥanbalite contemporary Ibn Taymiyyah, who, in his Sharḥ al ʿAqīdah al Aṣfahāniyyah (Explanation of the Creed of Aṣfahān) prefers a form of cosmological reasoning to demonstrate God’s existence.47 When presenting his viewpoint, Aquinas states: The third way is based on what need not be and on what must be, and runs as follows. Some of the things we come across can be but need not be, for we find them being generated and destroyed, thus sometimes in being and sometimes not. Now everything cannot be like this, for a thing that need not be was once not; and if everything need not be, once upon a time there was nothing. But if that were true there would be nothing even now, because something that does not exist can only begin to exist through something that already exists. If nothing was in being nothing could begin to be, and nothing would be in being now, which is clearly false. Not everything then is the sort of thing that need not be; some things must be and these may or may not owe this necessity to something else. But just as we proved that a series of agent [efficient] causes can’t go on forever, so also a series of things which must be and owe this to other things. So we are forced to postulate something which of itself must be, owing this to nothing outside itself, but being itself the cause that other things must be. And this is what everyone calls God.48

Aquinas’s argument, which can also be seen as an argument from contingency, can be summarised as follows: Contingent things are things which could be or not be. We see things around us which are contingent. If something can be or not be, it is not deemed necessary, but instead is caused by something else. There cannot be a world of only contingent things, as the explanation of such things lies with something other than itself. We cannot have an infinite series of contingent things, just as we cannot have an infinite series of causes. For this reason, we must have something self-sufficient and necessary which must ‘be of itself’ and owe such being to nothing ‘outside of itself’. Like Ibn Rushd, Aquinas defines imkān or possibility strictly using generation and destructibility. Unlike Ibn Sīnā, however, Aquinas does not employ the form of Tarkīb argument which argues that anything with distinguishable and dependent parts cannot be necessary. It may be argued that the lack of this feature makes the argument more susceptible to the fallacy of composition. A possible way around this is the use of the concepts of potentiality and actuality. According to Aquinas, ‘potentiality is actualized only by something already in actuality’.49 It can be argued using this concept that since the universe is constantly changing, this is evidence of its potentiality. Anything potential must require an external cause or actualiser. This argument has been used to great effect by Christian apologists such as Edward Feser, who argues at length that God is the necessary actualiser.50 Notably, Feser employed this argument against Graham Oppy to good effect in an online discussion entitled Are There Any Good Arguments for God? During this discussion, Oppy could not devise a solution to the problem of potentialities becoming actualities.

One of the biggest advantages of Aquinas’s third way and those who have commented on it is that it allows us to gauge an entire stream of Western philosophical criticism. This is practically expedient, as it chiefly relates to anticipated objections and reactions to these kinds of arguments in dialectic discourse. On this point, Michael Augros summarises the three main objections to the third way by stating:

(1) the premise that “What is possible not to be at some time is not” appears to be unknowable, (2) when Aquinas says “if therefore all things (omnia) are possible not to be, at some time nothing was in things” he is guilty of the quantifier shift, and (3) granting the entire argument, Aquinas has no right to conclude the existence of anything other than matter, which one might well believe is a self-necessary being’.51

As seen in the previous chapter (and putting aside the validity or lack thereof of these interrogations), Ibn Sīnā does not present the argument in a way that makes it susceptible to these criticisms. For example, the proposition ‘what is possible not to be at some time is not’ does not feature at all in Ibn Sīnā’s version of the argument. The idea that ‘If therefore all things are possible not to be, at some time nothing was in things’ would not seem to be commensurate with Ibn Sīnā’s reasoning at all. As for the third point of criticism associated with materiality, Anthony Kenny states that ‘in order to show that the uncaused everlasting being must be God, he offers no proof, and we may ask why might it not be perpetual, indestructible matter?’.52 Al Ghazālī, however, anticipates this line of reasoning in his Tahāfut by asking: But as for [you philosophers], what is there to prevent you from [upholding] the doctrine of the materialists – namely, that the world is eternal, that it likewise has no cause and no maker, that only temporal events have a cause, that nobody in the world is originated and nobody annihilated, but [that] which occurs temporally is but forms and accidents?53 Ibn Sīnā deals with the contention of materiality in considerable detail, as shown through the composition argument. As mentioned hitherto, Ibn Sīnā has a separate argument of composition responding to this in order to show that the uncaused and everlasting being must be God. This acts as a necessary corollary to his argument for an uncaused causer. Furthermore, this would indicate the impossibility of something which has a material form as being a necessary existence. It would seem therefore that the three major objections found in Western philosophy against Aquinas’s Third Way – as mentioned by Augros above – are nonfactors when returning to Ibn Sīnā’s original arguments.

Ibn Taymiyyah –who prefers the argument to be made cosmologically – makes a clear distinction between a ‘part’ and an ‘attribute’. According to Ibn Taymiyyah, a part is something that is materially added or removed from a substratum, like planks of wood that a whole [ship] depends upon for its existence.54 Anything that is constituted by material parts like these must be composed or generated. Considering that the universe – or a multiverse – would fit this description, it would follow that a universe or multiverse would be composed or generated. Ibn Taymiyyah goes on to argue that the attributes of God are talāzumī (necessary) such that they cannot be imagined in another way.55 Although Ibn Taymiyyah himself does not endorse it in ontological terms, nevertheless he renders the composition argument possible in terms similar to Ibn Sīnā’s set theory type formulations explored in the previous chapter. These medieval discussions have already been a staple part of the atheist versus theist discourse, especially with the rise of agnosticism in recent years. Ironically, it was Anthony Kenny himself – who was acting in his capacity as a moderator – challenged Richard Dawkins with this line of thinking. This was during the latter’s debate with Rowan Williams – the Archbishop of Canterbury at that time – regarding Dawkins’ argument of a complex God, where the example of an electric razor and a cutthroat razor was invoked. On this point, Kenny stated that ‘the cutthroat razor is simpler in design but has more complex powers than the electric razor’, after which Dawkins had no response but to plead ignorance, thereby demonstrating the effectiveness of this line of reasoning.56 One can see how making mereological distinctions – like those hitherto explored – is an essential part of making the argument from composition. If one defines a ‘part’ in a way that includes ‘attributes’, the apologetic mission may halt at a point of deism, as the affirmation of Godly attributes that are known through revelation becomes both redundant and superfluous. Many arguments for God’s existence can be quite complicated for an audience consisting of laypeople. They can be off-putting in pastoral settings, where the common folk just want to resolve atheistic doubts. In the Islamic golden age, the traditionalist Ibn Taymiyyah was most disapproving of employing overly complicated arguments when addressing lay people. He argued that they should be limited to an audience with an analytic background, interestingly pointing out that for ‘some people, every time the proof is more explicated and detailed, with more logical premises, it was more useful to them…and with this kind of person, one should use a detailed Kalām approach or other such analytic approach which they would be used to’.57 In another work entitled as Mas’alah Ḥudūth al-ʿĀlam (The Issue of the Beginning of the World), he states that the best way to convince a layperson of God is to use the most basic reasoning possible. Commenting on one of the Qur’an’s many rhetorical questions, ‘Or were they created from nothing or are they their own creators?’ (Qur’an, 52:35), Ibn Taymiyyah states: This categorisation is the easiest and clearest way that one can reason the existence of a creator with the most minimal amount of introspection. This is because the slave knows that he once did not exist, and that he came into existence after he did not exist…He also knows that he did not create himself or bring himself into existence, and this is known as a matter of critical certainty… He also knows that his creation could not be without a creator, and that there is no cause without effect… If he knows this, then he will know how to reason the existence of the heavens and the Earth.58

Al-Ghazālī makes a similar observation in his exegesis of the same verse referenced by Ibn Taymiyyah.59 Introducing the doubter to simple questions such as ‘do you think it is possible that the universe could come from nothing?’ or ‘is the universe dependent or independent?’ are useful analytical starting points. Arguing for the impossibility of causal or dependent things may be enough to arrive at the conclusion that an independent entity responsible for all other things in existence is necessary to explain anything.

Duns Scotus begins his argument in a similar manner to al-Ghazālī and Ibn Tamiyyah. However, he then explicates two types of arguments (with one being cosmological, and the other being ontological) in order to produce an inescapable conclusion. Rega Wood summarises Scotus’ argument in the following way: (1) Something can be produced.

(2) Everything that is produced is produced either by itself, nothing, or some other cause.

(3) Nothing can be produced from nothing.

(4) Nothing can produce itself. (5) Therefore, something is produced by another cause, which we will call a. (6) Either a is an uncaused cause or it is not. (7) If a is an uncaused cause, this suffices as proof. (8) If a is not an uncaused cause, then given steps (2-4), it must be produced by another cause, which we call b. (9) But it is impossible to have an infinite series of causes preceding each other. (10) Therefore, the procession must halt at some uncaused cause.60

Crucially, Scotus differentiates between accidentally ordered causes and essentially ordered causes. In the case of accidentally ordered causes, if one stops the other will not cease to exist. In the case of essentially ordered causes this is not the case. For example, a father may have a son, and the son may have another son. The grandfather may die, while both the son and the grandson will continue to exist. In this way, the son of the father and his own son can continue to exist despite the antecedent cause ceasing to exist.61 Like Aristotle and Ibn Sīnā, Scotus argues strongly against the possibility of an infinite regress of essentially ordered causes.62 Scotus’s argument has also been presented in the following manner: (1) If an independent entity (call it a) can fail to exist, then something incompatible with it (call it b) can exist, for one of two contradictories is always true.

(2) Something incompatible with an independent entity cannot exist because everything that exists is either independent or dependent (i.e. from itself or from another). (3) But b cannot be independent, since [if b is possible and independent], then it actually exists independently – from the third conclusion (conclusion 4 in De primo principio). (4) And if b is independent, then its possibility implies its actual existence, and its possibility would imply that two logically incompatible beings would actually exist. (5) b cannot be dependent. (6) For nothing incompatible with something that exists can receive being from a cause, unless it receives from its cause a more potent being than that which it is incompatible with – [in this case a], (7) But a is uncaused, and hence it is more potent than any being with a cause, since something caused owes its being to another entity. (8) [Therefore, b cannot come into existence as a dependent being, since it could not receive more potent being than a from any cause]. (9) Therefore, b cannot exist, and a cannot fail to exist.63

To transform this into a pure argument from necessity, one needs to only transform references from ‘caused things’ into ‘dependent things’. Expressed in one line, the argument against infinite regress is simply that ‘there cannot be an infinite regress of dependent things’. One may also effectively make a causation argument upon a parallel track – a suggestion made by Ibn Ṭufayl.64 Doing so will have the net effect of rendering the argumentative target smaller and more difficult to attack for interlocutors in polemical or apologetic settings. As it will be covered in much more detail later, the objections relating to causation would simply be irrelevant to at least one form of the argument. In this chapter, we have seen many of the sophisticated methods of demonstration employed by theologians and philosophers. In particular we have analysed the arguments of al-Ghazālī, Ibn Rushd (Averroes), Ibn Taymiyyah, Aquinas, and Duns Scotus. This is by no means an exhaustive list of important contributors or detractors to the argument put forward by Ibn Sīnā. Other theologians, such as al-Ṭūsī, Fakhr al Dīn al-Rāzī, al-Masʿūdī, as well as many in the Muʿtazalite tradition have written commentaries on the The Pointers or have addressed Ibn Sīnā’s Burhān argument in length. Many of these works are already available in the English language. Conspicuously absent from this chapter is also Maimonides, through whom Aquinas was introduced to Ibn Sīnā’s argument. We have also been able to establish that objections expressed against Aquinas’s contingency argument are not applicable to Ibn Sīnā’s particular version. Moreover, the importance of defining the modal categories has been made clear to us by almost all of the thinkers we have analysed. That being said, the fact that genii from varied religious traditions agree with at least one rendition of this argument speaks to the immense explanatory scope that such an argument has.

Betty may not be convinced that the universe is contingent. Using Aquinas’s lines of argumentation, one may be able to suggest to Betty that the universe is contingent because it is in constant change. Anything in constant change is potential and cannot be purely actual. Anything which is potential requires an outside cause. Some of the trauma Betty had suffered had been at the hands of her own family members. It may be that Betty may feel intimidated or distrustful of religious authority figures. It may also be that religion represents a greater authority that Betty disdains. Of course, it may be that Betty is genuinely not convinced with the arguments or that she doesn’t want to be convinced by any argument that leads to religious living. It is impossible to assess Betty’s intentions and psychoanalytic state. When interacting with Betty, I have found that the best way to be is as authentic as possible. To create a positive relationship with Betty.